Johansen cointegration test: which of 5 models? I have 4 variables (GDP, export, industrial production). I found that the variables are nonstationary in level but stationary in their first differences.
I conducted the Johansen test and got this result. I must choose between the five models. The appropriate model is chosen according to the information criteria. The problem is that I can't make an interpretation of the result that I have.
 A: It may not be advisable to choose the model just by using the information criteria; subject-matter knowledge can be very helpful. Sometimes you can rule out some of the candidate models because their shape clearly clashes with reality (despite good in-sample fit which may be due to chance). In your case, consulting macroeconomic theory could be worthwile.
However, if you are determined to choose the model using the information criteria, you should look for the lowest value of AIC and BIC -- the ones marked by asterisks in the table (model 3 or 4 for AIC and model 5 for BIC). If you are going to use the model for prediction, you may trust AIC over BIC while if your modelling is explanatory, you may trust BIC over AIC. Note that AIC is equally favourable to model 3 and model 4. Apart from subject-matter argumentation (which may be quite important), this is not really a problem -- both models are equally likely to be the best, so you will not win or lose much by choosing one over the other.
